Agricultural crop zoning with mixed integer programming for aromatic coconut in Ratchaburi Province

Abstract

Nowadays, the main concern in agricultural production is to balance the demand and supply to avoid the supply-over-demand problem. Agricultural zoning is a preferred solution to the problem, allowing farmers and policymakers to choose the most suitable economic crop for a particular area. The land suitability, factory locations and distances, profits, and crop growth costs must be considered when choosing optimal areas. For optimization problems, Mixed Integer Programming (MIP) is one of the most common mathematical optimization tools. In this study, we employed a solver called the Gurobi optimizer to find the target areas for the given crop. In this multi-criteria optimization problem, the goal is to maximize the land suitability of the desired crop, maximize profit, minimize the switching between the current and the new crop, and minimize the distance from the areas to the nearest market. The total amount of crop production should not exceed the capacity of the factories and the market demand. We apply the models to the real-world problem by assuming that the desired crop is Sweet Young Coconut (Aromatic Coconut) in Ratchaburi province. Land suitability of crops, economic inputs, and market locations for this study are obtained from the Agri-Map: Agricultural Map for Adaptive Management. We test the efficiency of the models with two scenarios: with and without factory purchasing capacity limitation. The results show that the model with factory purchasing capacity constraints is applicable for choosing the most suitable target areas to balance the crop supply with the market demand.